PxA Labs. Research and engineering for intelligent systems.

PxA Labs. An independent research collective from India building open, measurable intelligent systems, from safety monitors for fine-tuned LLMs and accent-robust speech recognition to interpretable genomics and real-time adaptive optics. Every result is reported against a fair baseline, including what did not work.

Research areas

Models, datasets and demos

ProjectOn the Hub
SafetyPulse: catches safety drift during fine-tuning, repairs it, audits the repairModels: tiny-aya-global-safetypulse-repaired · NF4 4-bit · LoRA adapters (alpaca, alpacacur, gsm8k, none)
Datasets: eval · finetune-data · pairs · probe-bundle · logs
IndiVoice-DeepASR: Whisper + LoRA for Indian-accented speechModel: whisper-indian-lora
AutoMaintainer: an always-on autonomous AI software engineering teamSpace: AutoMaintainer

Code for every project is on GitHub.

Connect

Website · GitHub · LinkedIn · Instagram · Discord · pxalabs@gmail.com